Project Summary Genome-wide association studies (GWAS) have been successful in detecting thousands of variants that are associated with human diseases and complex traits. However, for only a handful of these GWAS risk loci are the underlying biological mechanisms known. In particular, most GWAS variants lie in non-coding regions of genome that makes their biological mechanisms difficult to determine. Detecting the causal variants can help to understand the genetic architecture of disease and the biological mechanisms of non-coding GWAS risk loci. One way to determine the biological mechanism is to detect causal variants in molecular phenotypes such as gene expression, histone marks, and DNA methylation. This proposal aims to leverage molecular QTL to improve our understanding of the genetic architecture of disease and complex traits. The first aim is to develop a method to construct maximally disease-informative annotations from eQTL, hQTL, sQTL, and meQTL. The second aim is to prioritize gene sets that are important for common disease using common variants, making use of the disease-informative annotations from the first aim. The third aim is to understand the genetic architecture of trans-QTL. Ultimately, this proposal will create 1) a set of functional annotations for eQTL, hQTL, sQTL, and meQTL that are maximally enriched for disease heritability, 2) quantify what fraction of molecular QTL that are enriched for disease heritability are QTL-specific or shared among different molecular QTL, 3) prioritize disease-relevant gene sets, and 4) elucidate the genetic architecture of trans-QTL and their connection to disease heritability.